In recent years, the massive accumulation of health and medical data driven by the development of medical informatization has propelled the medical field into the era of big data. The construction and application of disease-specific databases have gradually become an important way to improve clinical diagnosis and treatment levels and promote the development of clinical research. This paper aims to provide reference for the construction of disease-specific databases in the future by reviewing relevant literature, summarizing key parts, technologies and the main applications of disease-specific databases, and proposing future development suggestions.
The practical nature of medicine and the individuality of diseases have led to high misdiagnosis rates, and clinical decision support system (CDSS) based on artificial intelligence and medical big data can provide decision support for doctors in the diagnosis of diseases, which has become an important means of solving such problems, and has now also achieved certain results. However, despite the potential advantages of CDSS in improving the accuracy and efficiency of medical decision making, there are a series of problems in its implementation that may affect the reliability, usability and safety of CDSS. This paper summarizes and analyses the current status of the application of diagnostic CDSS, the challenges it faces, and the future development trend, with a view to providing reference for the development of CDSS towards intelligence and knowledge in China.
Objective: To establish a special disease database based on clinical diagnosis and treatment data of pediatric allergic diseases so as to provide data support for multidisciplinary diagnosis and treatment and maximize the role of information technology in promoting the improvement of clinical diagnosis and treatment level and medical research progress. Methods: Based on the clinical diagnosis and treatment data of children with allergic diseases in Shanghai Children’s Hospital from January 2013 to July 2018, the clinical data was uniformly cleaned and stored using the extract-transform-load technology. The structured indicators in the original business system were mapped and standardized, while the unstructured content was manually labelled. Natural language processing technology was then applied for post-structured governance to create a specialized disease data model, which was used to build a dedicated pediatric allergic disease database. Results & Conclusion: The special disease database consists of 6 thematic data modules, including 16 items and 60 fields. We have achieved the collection and standardization of clinical diagnosis and treatment data for 333 029 children with allergic diseases, and further explored the distribution of diseases. The establishment of a special database for pediatric allergic diseases has enabled the storage, mining, and analysis of massive clinical data in the real world, providing support for the subsequent expansion of database based specialized management and decision-making assistance.